Freeform season chatter and light commentary

[vox_mundi]

[Steven]

I built a Google Site with an overview of my Dropbox graphs and spreadsheets:

https://sites.google.com/view/arctic-sea-ice

Meanwhile the website includes several Antarctic sea ice graphs too, so it needs a new name.  The old name of the website ("Arctic sea ice calculations and graphs") was always meant to be a placeholder anyway.

The best name I could come up with so far is "Polar Python", since the website uses python scripts, but I'm not sure if it's a good name since I'm not a native speaker.

The address of the website  (.../arctic-sea-ice/) was also meant to be a placeholder and probably should be changed too.  But that would break all existing links to it...

[oren]

You could also rename it to Polar sea ice calculations and graphs, or just Sea ice calculations and graphs.

[Supak]

Kalshi has added prediction markets for the 2024 acrtic max and min:

https://kalshi.com/markets/arcticicemax/arctic-sea-ice-max-extent#arcticicemax-24apr01

https://kalshi.com/markets/arcticicemin/arctic-sea-ice-min-extent#arcticicemin-24oct01

[Steven]

You could also rename it to Polar sea ice calculations and graphs, or just Sea ice calculations and graphs.

I never really liked that name:  it's long and sounds too similar to the name of Nico Sun's site (Cryosphere computing) and Neven's site (Arctic sea ice graphs). 

For now I'm keeping the new name, although it may still be changed if another idea comes up.  Also keeping the old website address for now, but it will probably change eventually since I'm planning to replace the Google Site by a proper website at some point in the future.

[kassy]

Polar Python does cover both suggestions and is easy to remember and it describes what the site does in a short way. It´s a good name.

[Supak]

More than 15 million km2 for the max is selling for 46 cents (to win a dollar):

https://kalshi.com/markets/arcticicemax/arctic-sea-ice-max-extent

[jdallen]

I added the monthly extent value for January 2024 into my long-term plot where I calculate the extent anomalies from 1979 up to now.

<snip>

The slope of the long-term linear trend line became flatter by two digits (-0.0523 → -0.0521).

See attached graph, now extended to 2029, a new five-year period has begun. Time is flying. Click to enlarge it.

Makes me think of a flat rock skipping across a pond.

[dnem]

jdallen said in the freezing thread: "Ice numbers have been bouncing up and down like a superball the last month, driven by heat breaking north primarily from the Atlantic."

This makes me think of the spinning top analogy for complex systems behavior on the verge of a change to a new equilibrium state (the top wobbles before it topples over).

Is there a good way to quantify if seasons have become more "bumpy" (less monotonic) over time?

[Glen Koehler]

     Related question - The decline trend for Extent and Area September minimums has flatlined since ca. 2007.  Thickness decline has continued but slowly.  The decline slope for Volume (a function of Area x Thickness) decline has been essentially flat since ca. 2010.  The quantitative situation appears relatively stable, but precariously so.


     At the same time, more qualitative factors such as ice pack integrity, mobility, and concentration seem to show a decline not reflected in the Ext/Area/Vol stats.  For example, ice flow in the CAA appears to be a possible harbinger of an imminent change in the equilibrium state of the ASI.  If what used to be the pile of thick ridged ice at the CAB-CAA border becomes an open gate for southbound extinction export of ASI, that would be a fundamental change in the ASI equilibrium.   

     Is there a consensus for metrics that track the qualitative condition of the ASI? 

     The standard metrics show stability, but the ASI condition over recent years seems to heading for another round of rapid decline parallel to the early 2000s when MYI got decimated before anybody could explain what was happening.  The current situation feels like the quiet before another storm.  The global average surface temperature anomaly vs. 1850-1900 is getting perilously close to the estimated +1.7C threshold for the September ASI minimum to go below 1M km2 and <15% of the early satellite period baseline.  Unlike more incremental impacts of climate change, the melting point of ice provides a discrete binary and relatable indicator for progressive warming.   

     But subjective feelings are not a valid assessment tool.  Are there objective numbers that tell a more complete story about the status of ASI than the relatively stable Ext/Area/Vol measures?

[gerontocrat]

This makes me think of the spinning top analogy for complex systems behavior on the verge of a change to a new equilibrium state (the top wobbles before it topples over).

Is there a good way to quantify if seasons have become more "bumpy" (less monotonic) over time?

Maybe, or maybe just a function of an ever-increasing gap between maximum and minimum

The graph attached is the sum of daily Arctic sea ice gains plus absolute value of losses year by year from 2000 to end 2023, which indicates greater activity as the years go by.

The linear trend is a gain of over 100,000 km2 per year, but the R2 value is a weak 0.32.

Also note the peaks in 2007, 2012, 2016 & 2020 which correspond to years of low sea ice minima.

The 2nd graph attached also shows greater values of freeze and melt over the years.

It might be owrth looking at wobbles during the couple of months around the maximum and the minimum

[Glen Koehler]

       FWIW - Skeptical Science rebuttal to the "ASI has returned to normal" claim includes this chart which extends the ASI Extent chart back to 1953.
 
       The chart ends at 2010, but the increasing height of the annual max to min oscillations described by Gero's graph are already apparent in this one too.  The increased swings are due to the summer/fall min values declining faster than the winter/spring max values. 
 
       This fits with warmer summers causing higher melt and less ice for the September minimums.  But then the negative feedback of faster heat loss and refreeze of larger areas of open water during the winter refreeze season (i.e. the "Slow Transition" process) results in a slower decline of the early spring annual maximum Extent.


https://skepticalscience.com/Has-Arctic-sea-ice-recovered-intermediate.htm

[The Walrus]

This makes me think of the spinning top analogy for complex systems behavior on the verge of a change to a new equilibrium state (the top wobbles before it topples over).

Is there a good way to quantify if seasons have become more "bumpy" (less monotonic) over time?

Maybe, or maybe just a function of an ever-increasing gap between maximum and minimum

The graph attached is the sum of daily Arctic sea ice gains plus absolute value of losses year by year from 2000 to end 2023, which indicates greater activity as the years go by.

The linear trend is a gain of over 100,000 km2 per year, but the R2 value is a weak 0.32.

Also note the peaks in 2007, 2012, 2016 & 2020 which correspond to years of low sea ice minima.

The 2nd graph attached also shows greater values of freeze and melt over the years.

It might be owrth looking at wobbles during the couple of months around the maximum and the minimum

More like dnem stated, a system equilibrium change occurred around 2005.  The rates have changed since then.

[dnem]

This makes me think of the spinning top analogy for complex systems behavior on the verge of a change to a new equilibrium state (the top wobbles before it topples over).

Is there a good way to quantify if seasons have become more "bumpy" (less monotonic) over time?

Maybe, or maybe just a function of an ever-increasing gap between maximum and minimum

Just spitballing here and don't have the time or probably the chops to do it myself. Fit an "idealized" monotonically increasing curve to each freeze season from min to max that minimizes either the absolute value of the deviations or the square of the deviations. Compare the sum of the absolute values or the squares to quantify how "bumpy" each freezing season is. No idea if this makes sense! (freeform chatter!). Or maybe truncate the data in some way to get around the "ever-increasing gap" issue: "last 10 million sq k of gain" or something.

[The Walrus]

This makes me think of the spinning top analogy for complex systems behavior on the verge of a change to a new equilibrium state (the top wobbles before it topples over).

Is there a good way to quantify if seasons have become more "bumpy" (less monotonic) over time?

Maybe, or maybe just a function of an ever-increasing gap between maximum and minimum

There was an increasing gap between maximum and minimum.  But a regime change in 2007 resulted in a large jump (>1 M km2) and then a decreasing gap.  Perhaps it is just decreasing to meet up with the previous [increasing] trend line.

[oren]

I guess trend lines are in the eye of the beholder. But eventually with freezing seasons being weaker and some peripheral seas refusing to freeze, and with the inner CAB ice thicker and resilient, the amplitude between max and min must decrease.

[Renerpho]

I guess trend lines are in the eye of the beholder. But eventually with freezing seasons being weaker and some peripheral seas refusing to freeze, and with the inner CAB ice thicker and resilient, the amplitude between max and min must decrease.

You'd think so. Of course the amplitude would go to zero for an Arctic that remains ice-free all year, but that's a long way off (if it actually happens), and there is no reason to assume that we approach zero in a straight line. In the mean time, we are melting areas that haven't seen any melt before, and those areas also refreeze easily, at least for now. That should temporarily increase the amplitude between max and min, until some of the peripheral seas remain ice-free all year. Then the more resilient CAB ice probably becomes the dominant factor, and the amplitude should indeed decrease again.

For once, Walrus' fit looks to be the more sensible one. (And I don't say that lightly, because I don't like it when they try to fit random step functions to sparse data!)

[The Walrus]

I guess trend lines are in the eye of the beholder. But eventually with freezing seasons being weaker and some peripheral seas refusing to freeze, and with the inner CAB ice thicker and resilient, the amplitude between max and min must decrease.

You'd think so. Of course the amplitude would go to zero for an Arctic that remains ice-free all year, but that's a long way off (if it actually happens), and there is no reason to assume that we approach zero in a straight line. In the mean time, we are melting areas that haven't seen any melt before, and those areas also refreeze easily, at least for now. That should temporarily increase the amplitude between max and min, until some of the peripheral seas remain ice-free all year. Then the more resilient CAB ice probably becomes the dominant factor, and the amplitude should indeed decrease again.

Walrus' fit looks to be the more sensible one. (And I don't say that lightly, because I don't like it when they try to fit random step functions to sparse data!)

I agree.  The question remains as to when will we reach that stage.  Perhaps we already did, back in 2007, as indicated in the graph.  Maybe the upward trend will return sometime in the future, and that stage is still well off. 

Then again, we may reach a point whereby the entire Arctic melts in the summer (or most of the open water), and then refreezes each winter, creating a large and recurring amplitude between max and min.

[Renerpho]

Maybe the upward trend will return sometime in the future, and that stage is still well off.

Implying a downward trend after the gap may be assuming too much. What is compelling about the step function fit is that you can fit a piecewise-constant function to that data, requiring few parameters (only one more -- 3 -- than a simple straight line fit to the entire data). I don't see enough evidence for any change before or after the gap.

[The Walrus]

Maybe the upward trend will return sometime in the future, and that stage is still well off.

Implying a downward trend after the gap may be assuming too much. What is compelling about the step function fit is that you can fit a piecewise-constant function to that data, requiring few parameters (only one more -- 3 -- than a simple straight line fit to the entire data). I don't see enough evidence for any change before or after the gap.

Quite possibly.  Neither trend line portrays the data very well.  The earlier trend line has a better fit simply due to a higher number of datapoints (28), than the more recent line (17).  In both cases the annual scatter overwhelms any fit.  All things considered, I would say each trend represents the data to a similar degree, albeit not very well.

[gerontocrat]

This makes me think of the spinning top analogy for complex systems behavior on the verge of a change to a new equilibrium state (the top wobbles before it topples over).

Is there a good way to quantify if seasons have become more "bumpy" (less monotonic) over time?

FAILURE on the "monotonic" thingummy

I thought - looks at the wobbles around the minimum and the maximum. Why? because if the ice is getting thinner, it is more mobile and can more easily melt and freeze again. Using part year periods might also avoid the trends appaearing when using all year data.
That was the idea.

So I produced two graphs, looking at the 2 months approximately before and after the minimum and the maximum.

The results are IMO pretty meaningless on the monotonic thingummy. But they do show wobbles are much higher around the minimum compared with around the maximum, which seems logical as the ice is at its thinnest at the minimum.

[kassy]

And what if we split them into the 4 seasons?

[dnem]

I'm not surprised. There is no apparent increasing trend over the years in "the wobbles" indicated by simply eyeballing the graphs.

[Glen Koehler]

      Is there a consensus for metrics that track the qualitative condition of the ASI? 

     Science Citation Index would indicate if anybody has followed up on this 2013 study (too far behind at work and in life to track it down myself).  A 2013-2023 update of the exchange between CAB-Beaufort and CAA would be interesting to see.  I don't understand how the data charts in this paper connect to the conclusions stated in text.  And the results for QEI and McClure Strait seem contradictory.  I thought ice just moved south from the AO into the CAA but apparently there's more to it than that.   

   Howell, S. E. L., T. Wohlleben, M. Dabboor, C. Derksen, A. Komarov, and L. Pizzolato (2013), Recent changes in the exchange of sea ice between the Arctic Ocean and the Canadian Arctic Archipelago, J. Geophys. Res. Oceans,118, 3595–3607,doi:10.1002/jgrc.20265
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/jgrc.20265

Excerpt from the Abstract:
     "The presence of MYI in the CAA originating from the Arctic Ocean has been maintained by inflow at the QEI, which has increased since 2005. These recent increases in Arctic Ocean MYI inflow into the QEI can be attributed to increased open water area within the CAA that have provided more leeway for inflow to occur."


"Time series of Arctic Ocean-M'Clure Strait gate net total ice and multi-year ice area flux from (a) May to September and (b) October to November, 1997–2012. Positive and negative flux signs correspond to Arctic Ocean ice inflow and outflow, respectively."


"Time series of May to September net open water within the Western Parry Channel and May to September Arctic Ocean-M'Clure Strait net ice area flux, 1997–2012. Positive and negative flux signs correspond to Arctic Ocean ice inflow and outflow, respectively."


"Time series of Arctic Ocean-Queen Elizabeth Islands (North and South) gate net total ice and multi-year ice area flux from (a) May to September and (b) October to November 1997–2012. Positive and negative flux signs correspond to Arctic Ocean ice inflow and outflow, respectively."


"Time series of May to September net open water within the Queen Elizabeth Islands and May to September Arctic Ocean-Queen Elizabeth Islands (North and South) net ice area flux, 1997–2012. Positive and negative flux signs correspond to Arctic Ocean ice inflow and outflow, respectively."

Excerpts from Conclusions
     "We utilized 16 years of RADARSAT imagery to estimate the sea ice area flux between the Arctic Ocean and the CAA at the M'Clure Strait and QEI (north and south) exchange gates for the months of May to November from 1997 to 2012.  Over this period, the M'Clure Strait ice area flux was −1 × 10^3 km2 indicating net outflow to the Arctic Ocean but it is important to note that the interannual variability is high (i.e., ±21 × 10^3 km2).  An examination of seasonal differences revealed that while Arctic Ocean ice inflow primarily occurred during the summer months (May to September) outflow dominates in the fall months (October to November). The QEI gates experienced a mean flux of +3 × 10^3 km2 from August to September with negligible ice exchange from May to July and October to November.  These results indicate that Arctic Ocean ice inflow at the M'Clure Strait and QEI exchange gates during August and September play an important role in determining the annual sea ice minimum within the CAA.

     Since 2007, Arctic Ocean ice inflow at the M'Clure Strait gate during the summer months has reduced considerably. We attribute the decrease to the increased frequency and location of high SLP anomalies over the Beaufort Sea and Canadian Basin that disconnect the Arctic Ocean polar pack ice from ice within the Western Parry Channel. While Arctic Ocean ice inflow has decreased at the M'Clure Strait in recent years, the QEI gates have experienced increases since 2005. The increase at the QEI gates can be attributed to increased open water within the QEI (and the channels to the south of the QEI) providing more leeway for atmospherically driven Arctic Ocean ice import to occur."

     "Our analysis points out that despite reduced Arctic Ocean ice inflow into the CAA at the M'Clure Strait, ice inflow at the QEI has continued, thus maintaining the presence of Arctic Ocean MYI within the CAA. Although younger and thinner [Maslanik et al., 2011], this MYI has continued to flow southward into the channels of the Northwest Passage."

     "Sea ice conditions within the CAA during September have begun to decrease considerably in recent years, and the lack of Arctic Ocean ice inflow at M'Clure Strait has likely played a role."

Legend for attached graph:  "Comparison of 1997–2002 monthly May to November ice area flux at (a) the Arctic Ocean-M'Clure Strait and (b) the Arctic Ocean-Queen Elizabeth Islands (North and South) exchange gates between Kwok [2006] and this study. Positive and negative flux signs correspond to Arctic Ocean ice inflow and outflow, respectively.

     To the extent that I understand this material, it seems that the May - November flux between the AO and Queen Elizabeth Islands could be useful as a monitor of ASI export dynamics, parallel to the Farm Strait monitoring.  Especially if thinning ice along the southern border of the CAB and Beaufort, and ice loss in the CAA, is causing there to be increased export activity from the Arctic Ocean into the CAA.

[Glen Koehler]

     Anybody else have a tickle in their throat or an itch in their armpit suggesting that the 2024 melt season is going to be a real doozy?  With the 2023 global surface temperature far exceeding any previous year in the record book, and 2024 likely to be as warm if not warmer, and Antarctic summer sea ice once again at near record low for a third year in a row, something has to give in the Arctic. 

     The Arctic Exent and Area trendlines have been flat for over a decade.  My itch tells me that the cumulative qualitative changes in the ASI pack are about to break out and be revealed in the Ext/Area 2024 summer melt stats.  Pure guesswork of course, but you can't heat an ice cube forever.  Eventually, the ice cube melts.

[The Walrus]

     Anybody else have a tickle in their throat or an itch in their armpit suggesting that the 2024 melt season is going to be a real doozy.? With 2023 global surface temperature far exceeding any previous year in the record book, and 2024 likely to be as warm if not warmer, and Antarctic summer sea ice once again at near record low for a third year in a row, something has to give in the Arctic. 

     The Arctic Exent and Area trendlines have been flat for over a decade.  My itch tells me that the cumulative qualitative changes in the ASI pack are about to break out and be revealed in the Ext/Area 2024 summer melt stats. Pure guesswork of course, but you can't heat an ice cube forever.  Eventually the ice cube melts.

My instincts tell me to never go against the trend.  I am not anticipating anything, but who knows?

[Glen Koehler]

     Hi Walrus - Fair enough, but which trend? The last 12 years or the longer decline since 1978?

[The Walrus]

     Hi Walrus - Fair enough, but which trend? The last 12 years or the longer decline since 1978?

Since we were talking one year, 2024, I would say short term.  All gero’s plot show that 2023 is well above the long term decline, so they may not be as relevant in the short term.

[VeliAlbertKallio]

Frozen Isthmuses' Protection Campaign of the Arctic and North Atlantic Oceans (FIPC) and Sea Research Society (SRS) have published in 2024 an update to assessment of Arctic Compendium on Arctic Sea Ice manipulation practises. Links to the update and compendiums assessments here:

https://www.academia.edu/116595854/SEA_ICE_GROWTH_MANAGEMENT_further_information_to_Arctic_compendium

https://www.academia.edu/116573805/Sea_Ice_Growth_Management_an_assessment_in_Compendium_of_interventions_to_slow_down_halt_and_reverse_the_effects_of_climate_change_in_the_Arctic_and_northern_regions_pp_36_37

https://www.academia.edu/116573431/Iceberg_Management_an_assessment_in_Compendium_of_interventions_to_slow_down_halt_and_reverse_the_effects_of_climate_change_in_the_Arctic_and_northern_regions_pp_23_24

https://www.academia.edu/4302181/Rowe_Mark_and_Kallio_Veli_A_Can_space_mirrors_save_the_planet_As_it_becomes_ever_clearer_that_simply_cutting_back_on_carbon_emissions_isn_t_going_to_save_the_poles_

[VeliAlbertKallio]

A new paper draws light on depression system formation over Atlantic Ocean which then can veer north-east and can drive winds on the Fram Straight and the Barents Sea before they fizzle out. Hopefully this new information can be integrated to North Atlantic weather facing the Arctic.
https://agupubs.onlinelibrary.wiley.com/doi/epdf/10.1029/2023GL106872